Unnatural pyrimidine nucleosides showing antitumor activities, such as 5-fluoro-2'-deoxyuridine and 5-trifluoromethyl-2'-deoxyuridine, have already known to have strong in vitro activities [Cancer Research, 18, 335 (1958); 22, 815 (1962); 28, 2529 (1968); Proceedings of the Society for Experimental Biology and Medicine, 97, 470 (1958)].
However, these compounds are known to be promptly decomposed and inactivated in vitro by pyrimidine nucleoside phosphorylase which are found in the liver, the small intestine and the like [Cancer Research, 32, 247 (1972); Japanese Journal of Cancer and Chemotherapy, 8, 262 (1981); 8, 1548 (1981)], so that none of them have been found to bring about satisfactory clinical antitumor effects [Cancer Chemotherapy Reports Part 1, 55, 205 (1971); Physicians' Desk Reference, 32, 1387 (1978)].
With a view to preventing the inactivation, research have hence been conducted to develop inhibitors for pyrimidine nucleoside phosphorylase, and some strong inhibitors have been reported. Incidentally, there are two types of pyrimidine nucleoside phosphorylases, that is, uridine phosphorylase and thymidine phosphorylase. It has been reported that uridine phosphorylase is a primary enzyme in selenodonts such as mice and rats while thymidine phosphorylase is a principal enzyme in human and the like [Japanese Journal of Cancer and Chemotherapy, 8, 262 (1981)]. Potentiating the antitumor effects in human therefore requires an inhibitor for thymidine phosphorylase rather than an inhibitor for uridine phosphorylase.
However, a great majority of inhibitors which have been reported to date selectively exhibit inhibitory activities against uridine phosphorylase and show practically no activities against thymidine phosphorylase. Reported to date as exceptions, in other words, as inhibitors for thymidine phosphorylase are 6-amino-5-bromouracil and 6-aminothymine [Biochemical Pharmacology, 29, 1059 (1980)], 6-amino-5-chloro-uracil and 3-cyano-2,6-dihydroxypyridine [Japanese Patent Application Laid-Open (Kokai) No. SHO 63-250324], acyclothymidine [Japanese Patent Application Laid-Open (Kokai) No. HEI 5-213761], and the like. Their inhibitory activities are however not sufficient.
Further, human thymidine phosphorylase has recently been found to be the same as PD-ECGF (Platelet Derived Endothelial Cell Growth Factor) which is a human endogenous angiogenic factor [Nature, 356, 668 (1992)]. Accordingly, a thymidine phosphorylase inhibitor can inhibit angiogenesis which is closely associated with malignancy of pathomas such as solid tumors, rheumatism and diabetic retinopathy, and is useful as a therapeutic for these diseases.
In addition, 5-trifluoromethyl-2'-deoxyuridine also has antiviral activities and is used as an eye drop for herpetic keratitis [Science, 145 (3632), 585 (1964); American Journal of Ophthalmology, 73, 932 (1972)]. Phosphorylase inhibitors are also expected to have utility as enhancers for antiviral activities and effects.
An object of the present invention is therefore to provide a novel compound which has excellent inhibitory effects on human-derived thymidine phosphorylase and is useful as an antitumor effect potentiator and an antitumor agent.
With the foregoing circumstances in view, the present inventors have proceeded with extensive research. As a result, it has been found that a uracil derivative represented by the below-described formula (1) has excellent inhibitor effects on human-derived thymidine phosphorylase, leading to the completion of the present invention.